Development of bispecific antibodies for selective stimulation of the CD95 death receptor on malignant and normal activated B-cells

Abstract

Treatment of oncological and autoimmune diseases still remains a considerable challenge to modern medicine. Current therapy options such as chemotherapy and radiotherapy in oncology and glucocorticoids or other conventional immunosuppressive drugs for autoimmune diseases are often of limited efficacy. They suffer from a lack of specificity, with respect to the eradication of tumour and immune cells, respectively. The use of monoclonal antibodies in oncology allowed in some instances specific targeting of tumour cells and thus, improved this situation in the last decades. Some antibodies, such as anti-CD20 antibody Rituximab, targeting normal and malignant B-cells, are meanwhile firmly established in oncological treatment regimes. Interestingly, over the last years the anti B-cell activity of Rituximab was used in addition in numerous studies for the treatment of autoimmune diseases. Many strategies have been developed to further improve therapeutic activity of monoclonal antibodies. One of them is the development of bispecific antibodies recognising target cell antigens and effector cells or effector molecules at the same time. The “bispecific concept” used in this work is the target cell restricted stimulation of the death receptor CD95 (APO-1/Fas) with bispecific antibodies as described by Jung et al. in 2000. Whereas initial experiments were performed using chemically hybridised bispecific antibodies with CD20×CD95 specificity, the goal of this thesis was the generation of recombinant antibodies within a suitable format. To this end, a recombinant CD20×CD95 molecule in the so called FabSc format, designated BS9520 was developed and characterised in various in vitro and in vivo assays. It was found that the capability of this antibody to suppress the growth of malignant B-cells in vitro and in vivo and to specifically deplete normal, activated B-cells from PBMC cultures was superior to that achieved with monospecific clinically established anti-CD20 antibodies including a newly developed third generation Fc-optimized CD20 antibody. Moreover, the bispecific antibody was the only reagent capable of significantly suppressing IgG production by activated B-cells in vitro. These findings imply that the bispecific CD95×CD20 antibody might become an attractive reagent for the treatment of B-cell malignancies as well as B-cell mediated autoimmune diseases. In the second part of this work a completely new approach for the stimulation of the CD95 death receptor on antigen specific B-cells was evaluated. For proof of principle, the CD20 part in the BS9520 antibody was replaced by antigen fragments (tetanus or diphtheria toxoid derivatives). We named these fusion proteins BS95TT and BS95DT, respectively. Such protein should bind to B-cell receptor via their toxoid part and with the CD95 antibody part to CD95 receptor on activated TTspecific B-cells. Thereby, apoptosis is induced only in an antigen-specific fashion. This approach could be used for the treatment of autoimmune diseases if the antigen inducing auto-antibody production is known (such as Myasthenia gravis, multiple sclerosis and others).